JPS589009A - Inspecting device for substrate - Google Patents

Abstract

PURPOSE:To decide arbitrary patterns automatically and accurately by processing the image information read by an image sensor with a microcomputer. CONSTITUTION:In a sensor part 100, the patterns on a substrate 1 are made into image information by means of a sensor 2 driven by a driving circuit 3 and a roller 7 driven by a motor 6 and this information is read into a CPU5 via an AD converter 4. The original picture which is read first is stored in a hard disc 12, and the position reference and inspection area of the original picture are assigned by the operation of a master CPU9. While viewing a part of the specimen on a display 14, the operator enlarges said patterns with a slave system 500 and erases pinholes alpha and pattern cuts beta, then the patterns are reduced and the defective parts are detected by arithmetic operations.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a board detection device that detects unnecessary portions by comparing a pattern on a printed circuit board with an original image.

Conventionally, human eyes have been used to judge the quality of printed circuit boards (printed patterns, etching patterns, 9 turns, zigzag turns, etc.).

Therefore, there are disadvantages such as it is very time consuming, judgment of pass/fail differs from person to person, and oversights may occur.

This invention reads an image using an image sensor as a substitute for the human eye, calculates and compares the data with a microcomputer, and determines whether the image is good or bad. It is an object of the present invention to provide a board inspection device that eliminates individual differences in criteria for determining defects on a board, improves productivity, and eliminates the need for a dedicated jig.

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the board inspection apparatus of the present invention will be described below with reference to FIG. In Fig. 1, 1 is the board to be inspected, 2 is the image sensor in the corner □ that reads the no + turn on the board 1, and the image sensor 2 is driven by the driver circuit 3. The analog output of the image sensor 2 is converted into a digital output by the A/D converter 4.The output of the A/D converter 4 is transferred to the CPU 5. It's like a trap. This CPU 5 receives commands from the master CPU 9 and controls the sensor driver 3, A/D converter 4, and Stellar pin motor 6, respectively.The stepping motor 6 controls the roller 7, thereby controlling the substrate. It is a trap to move 1. This stepping motor 6
The roller 7t is driven so as to move in units of 0μ.

Reference numeral 81 denotes a general arm 4-sinter 7-eighth path for connecting each CPU and transmitting signals such as commands and data. Thus, substrate 1, image 7 sensor 2%
Senna driver 3, A/D converter 4, CPU 5, stepping motor 6, roller 7, generator 2/#-4-
Mainter 7 Eighth Pass 8a Toyoshi Seven System 1
00 Ka I11 has been completed.

The master CPU 9 is the center of this system.
The PU issues operation commands for each CPU. Further, 10 is a floppy disk device containing operation commands and management programs for the entire system; this floppy disk device 10. Master CPU9,'/Eneralnoya-4-
Sinter 7 Eighth Pass 8b and K Yoshimasta System 2
It is composed of 00.

The disk system 300 includes the node disk device 12,
CPUII, Generalno parable? - It is composed of Spinter 7 Eighth Pass 8C and K, and has a CPU that runs a CPU II line hard disk 12, and the hard disk device is a node disk device with a storage capacity of 20 MB to store original drawings. belongs to.

13 is a CPU that controls the color graphics display 14 and printer 15; the color graphic display device 14 is for displaying defective areas (ToF); and the printer 15 is for printing out defective areas. It is a first-class printer. This C
PU13. A color graphic display 14, a printer 15, a general notebook interface path 8d, and a display system 400 are configured.

Furthermore, slave system 500Fi CPU16°17%
It is composed of special hardware 18, 19, a pose interface path 8, 8f, and a CPU 16 which compares the original image and the read arm turn. The CPU 17 is an additional CPU having exactly the same functions as the CPU 16 when the size of the board 1 is large and the amount of image comparison is large. The special hardware is a circuit for enlarging and reducing the read Noreen, and the C1* special hardware 19 is an extension circuit having exactly the same function as the special hardware 18.

Next, the operation of the substrate inspection apparatus of the present invention will be explained based on the flowchart shown in FIG. First, in step A, the power is turned on and the process moves to step BK. In step B, the program to be executed is transferred from the master system 200 to the slave system 500. The slave system 500 thereafter executes this program.

Next, the process moves from step B to step fcK, and it is determined whether a new original image is to be read or whether the original image has already been read.If the original image is to be read into the new nine, the process moves to step DK, and the sensor system 100 and disk system 30
0 reads the entire original image and stores it in the hard disk 12. Next, while viewing the original image on the color graphic display 14 of the display system 400, the original image is checked (position reference) and inspected in step E by operating keys on the master system 200. Specify the area.

This designation will be used as the position reference and comparison area reference for the comparison turns that will be input later. If the original image has already been read and stored on the hard disk 12, the above operation is not necessary.

Next, in step F, the entire system is initialized, and the system enters into operation for comparing the image with the original image.

First check the sensor system 1 on board 1
Load with 00 (Stella 7'G). Check l India is checked at least two places, and based on the read information, the slave system 5oo calculates the enlargement coefficient, rotation amount, and translation amount when comparing the image with the original image in step HK.

Next, the sensor system 100 reads a part of the main turn on the board 1 using the Step E, and the information is transferred to the slave system 500 using the Stella fJ.

The slave system 500 is loaded using special hardware 18 and enlarges and reduces the nine-arm turn in steps to inspect for whiskers and pinholes. If a defect is found here, it is displayed on the display system 400 in step L.

During this time, the disk system 300 prepares to transfer the original image for comparison with the next original image in parallel in step L, and transfers a part of the original image to the slave system 500 in step MK.

Next, the slave system 500 enters the deviation correction amount between Stella 7" (1st, the comparison board calculated earlier) 4 turns and the original image and compares the image with the original image in step N. If there is a mismatched part with the original image, In step P, the portion is displayed on display system 400.

Next, in step Q, it is determined whether the entire board has been read, and if it has not been read, the process moves to step RK, the stepping motor 6 is driven by the CPU 5, and the board 1t is read.
Move it by 60μ, and read the Noreen on the substrate 1 in the same way. Thereafter, this operation is repeated until the entire board 1 is read.

After completing the inspection of all AI turns on the substrate 1, the display system 400 prints out all indications of defective locations in step S.

Through the above operations, it is possible to inspect the whiskers, pinholes, and portions of the main turn on the substrate 1 that are different from the original image. Note that the resolution of this system is determined by the number of bits of the image sensor 2, which determines how large one pixel is, and the memory capacity of the hard disk 12 for storing pixels.

This system is used to enlarge and reduce images, detect whiskers of 2 pixels or less, pinholes, etc., and a specific example is shown in Figure 3. For three or more pixels, pixel comparison is performed.

Figure 3 shows one hole α of 2 pixels or less, and the arm turn cut β
3(a) to 3(6).
)) By enlarging Kt by 2 pixels, K, shibin hole α, and noda turn cut β are eliminated. Next, reduce it as shown in Figure 3 (Figure 3), and it becomes as shown in Figure 3 (0) when subtracting from Figure 3 (c) to Figure 3 (Figure 3). This Section 311(d)-t.
This means that only the open hole α and the no-turn cut β portion remained and could be detected.

Figure 4 shows a method for detecting whiskers r of 2 pixels or less.
If the image is reduced by two pixels as shown in the fourth axis from the figure (b), the image r becomes more powerful. Figure 4(e)
If you enlarge it as shown in 8 and subtract it from 411(a) in FIG. 4G), you will get k as shown in FIG. 4O).

In this Fig. 4 (Gi), only the part of the whisker r is gI4, and the detection becomes 1 and K.

In the above embodiment, the detection of defects in the main turns (printed nocturnals, etched nodules, silk noda turns, etc.) on the substrate 1 was explained, but it is also possible to read the hole diameter (through holes, etc.) and hole positions in the original image. ! R, N
It is also possible to automatically create a C tape.

As described above, 1. According to the board inspection device of the present invention, an image sensor is used to read the board pattern, the read data is calculated by a microcomputer, and it is compared with judgment criteria given by a program. Since the pass/fail judgment can be made by using a single judgment criterion, any four turns on the board can be inspected without contact, and there is no need for a special jig. It is possible to eliminate individual differences and improve productivity through automatic inspection.

[Brief explanation of drawings]

FIG. 1 is a pull diagram showing the configuration of an embodiment of the board inspection device of the present invention, FIG. 2 is a flowchart for explaining the operation of the board inspection device of the present invention, and FIGS. ! J(d) and FIGS. 4(a) to 4@) are explanatory views of the enlargement and reduction methods used in the substrate inspection apparatus of the present invention, respectively. DESCRIPTION OF SYMBOLS 1... Board, 2... Image sensor, 3... Sensor driver, 4... A/D converter, 5, 9, 11
゜13.16.17...CPU, 6...Stepping motor, 10...70 disc, 12...
Hard disk, 14... Color graphic disk, 15... Printer, 18.19... Special hardware, 100... Senna system, 200... Mask system, 300... Disk system, 400
...Display system, 500... Slates system. Note that the same reference numerals in the figures indicate the same or corresponding parts. Agent Shin Kuzuno −

Claims (1)

[Claims] an image sensor that reads the pattern on the board; an A/D converter that converts the analog output of the image sensor into a digital output; a stepping motor that moves the board in predetermined dimensional units; a first storage unit that stores the original image; a second storage means for storing operation commands and management programs for each part;
a first means for determining the quality of the information on the parable turn by comparing it with the information of the original picture stored in the storage means; the first means determines that the information on the /4 turn is defective; A board inspection apparatus comprising a second means for displaying or printing out a determination result when a determination is made.